Workshop Programme - School of Electrical Engineering and
2014 14
th
UK Workshop on Computational Intelligence (UKCI)
UKCI2014
8-10 September 2014
University of Bradford
General Chair: Professor Daniel Neagu
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
2014 14 th
UK Workshop on Computational Intelligence (UKCI)
Copyright and Reprint Permission:
Abstracting is permitted with credit to the source. Libraries are permitted to photocopy beyond the limit of U.S. copyright law for private use of patrons those articles in this volume that carry a code at the bottom of the first page, provided the per-copy fee indicated in the code is paid through Copyright Clearance Center, 222 Rosewood Drive,
Danvers, MA 01923. For reprint or republication permission, email to IEEE Copyrights
Manager at pubs-permissions@ieee.org. All rights reserved.
Copyright ©2014 by IEEE.
IEEE (Xplore Compliant) Catalog Number CFP1448L-ART
ISBN 978-1-4799-5538-1
IEEE (USB) Catalog Number CFP1448L-USB
ISBN 978-1-4799-5537-4
2
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
2014 14
th
UK Workshop on Computational
Intelligence (UKCI)
University of Bradford
Student Central Lecture Theatre SC0.51
Bradford, West Yorkshire, UK
8-10 September 2014
Editors
Daniel Neagu
Mariam Kiran
Paul Trundle
Proceedings:
IEEE (Xplore Compliant) Catalog Number CFP1448L-ART
ISBN 978-1-4799-5538-1
IEEE (USB) Catalog Number CFP1448L-USB
ISBN 978-1-4799-5537-4
3
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Contents
UK Workshop on Computational Intelligence (UKCI) ........................................... 3
0800 – 1200 Registration (Student Central Boardroom SC0.56) ................................ 10
0900 – 0930 Welcome Talk (Student Central Lecture Theatre SC0.51) ..................... 10
1100 – 1240 Session 1: Emerging and Applied Computational Intelligence ............... 10
1320 – 1500 Session 2: Advances and Applications in Data Mining .......................... 10
– 1700 Session 3: Advances & Applications in Signal and Image Processing .. 11
– 1730 Guided Walk to Bradford City Hall, Centenary Square, Bradford ........... 11
1730 – 1830 Lord Mayor: Welcome and Tour of Bradford City Hall ........................... 11
1830: Welcome Reception (Forster’s Bistro and Deli, Centenary Square, Bradford) . 11
1020 – 1200 Session 4: Fuzzy Sets and Fuzzy Systems ........................................... 12
1330 - 1510 Session 5: Advances and Applications of Evolutionary Computing and
1630 - 1730 UKCI Matters: Chair Professor Qiang Shen, Aberystwyth University ..... 13
UKCI2015 : Professor Richard Everson General Chair, University of Exeter ..................... 13
1830: UKCI2014 Dinner. Best Student Paper Award presentation (Midland Hotel) .... 13
September 2014 .................................................................................. 14
0900 - 1000 Session 6: Intelligent Technologies for Healthcare and Behaviour
1020 - 1140 Session 7: Intelligent Technologies for Healthcare and Behaviour
4
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Computational Intelligence Research for Impact in the Games Industry: Professor Peter
Deriving the most value out of local data: Mark Barrett, Open Data Lead for Leeds ..... 16
Pathfinding in partially explored games environments. The application of the A*
Evolution-In-Materio: Solving Function Optimization Problems Using Materials ......... 17
Integration Strategies for Toxicity Data from an Empirical Perspective ...................... 17
Nonlinear control of the underactuated hovercraft using the Derivative-free nonlinear
Integrate Classifier Diversity Evaluation to Feature Selection Based Classifier
Social Media Analysis for Product Safety using Text Mining and Sentiment Analysis 20
Kernel Learning Method for Distance-Based Classification of Categorical Data ........ 20
A New Weighting Scheme and Discriminative Approach for Information Retrieval in
Session 3: Advances and Applications in Signal and Image Processing ....................... 21
Adaptive Learning with Covariate Shift-Detection for Non-Stationary Environments .. 21
Search-guided Activity Signals Extraction in Application Service Management Control
Modeling Neural Plasticity in Echo State Networks for Time Series Prediction .......... 22
Neural Networks and Wavelet Transform in Waveform Approximation ...................... 22
A Fuzzy Image Congealing-based Handwritten Chinese Character Recognition and
Fuzzy Adaptive Parameter Control of a Late Acceptance Hyper-heuristic ................. 23
Fast Learning method of Interval Type-2 Fuzzy Neural Networks .............................. 24
5
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Refinement of Fuzzy Rule Weights with Particle Swarm Optimisation ....................... 24
Session 5: Advances and Applications of Evolutionary Computing and Optimization .... 24
A Benchmark Generator for Dynamic Multi-objective Optimization Problems ............ 24
PermGA Algorithm for a Sequential Optimal Space Filling DoE Framework .............. 25
Hybridisation of Decomposition and GRASP for Combinatorial Multiobjective
Dynamic Railway Junction Rescheduling using Population Based Ant Colony
Crowd Formal Modelling and Simulation: The Sa’yee Ritual ...................................... 26
Soft Morphological Filter Optimization Using a Genetic Algorithm for Noise Elimination
A Rule Based System for Diagnosing and Treating Chronic Heart Failure ................. 27
Computational Techniques for Identifying Networks of Interrelated Diseases ............ 27
Word Order Variation and String Similarity Algorithm to Reduce Pattern Scripting in
Automatic Image Annotation with Long Distance Spatial-Context .............................. 28
PCA-based Algorithmic Approximation of Crisp Target Sets ...................................... 28
Optimized Artificial Neural Network Using Differential Evolution for Prediction of RF
Power in VHF/UHF TV and GSM 900 Bands for Cognitive Radio Networks .............. 29
Truss Topology Optimization with Species Conserving Genetic Algorithm ................. 29
Improving Motion State Change Object Detection by Using Block Background Context
Session 6: Intelligent Technologies for Healthcare and Behaviour Analysis 1 ............... 31
An Efficient System For Preprocessing Confocal Corneal Images For Subsequent
Human Activity Classification Using A Single Accelerometer ..................................... 31
Session 7: Intelligent Technologies for Healthcare and Behaviour Analysis 2 ............... 32
Blood-Glucose Pattern Mining Algorithm for Decision Support in Diabetes
Integrated Microarray Analytics for the Discovery of Gene Signatures for Triple-
Prediction of Driver Fatigue: Approaches and Open Challenges ................................ 33
Using Computational Methods for the Prediction of Drug Vehicles ............................ 34
6
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Organising Committee
General Chair:
Professor Daniel Neagu
Programme Co-Chairs:
Professor Qiang Shen
Professor Yaochu Jin
Dr Daniel Kudenko
Publication Co-Chairs:
Dr Mariam Kiran
Dr Paul Trundle
International Liaison Co-Chairs:
Professor Gongde Guo
Dr Mohammed Azzeh
Dr Yousef Alqasrawi
Dr Urszula Markowska Kaczmar
Special Session Co-Chairs:
Professor Rami Qahwaji
Dr Waleed Al Nuaimy
Web Chair:
Luke Hampshaw
Reception Desk/Administration:
Aisyah Mat Jasin
Haruna Isah
Uyoyo Zino Edosio
Pritesh Mistry
Omololu Makinde
University of Bradford
Aberystwyth University
University of Surrey
University of York
University of Bradford
University of Bradford
Fujian Normal University, China
Applied Science University, Jordan
Applied Science University, Jordan
Wroclaw University of Technology, Poland
University of Bradford
University of Liverpool
University of Bradford
University of Bradford
University of Bradford
University of Bradford
University of Bradford
University of Bradford
7
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Programme Committee
Alexander Brownlee
Anna Kononova
Anna Palczewska
University of Stirling
Heriot-Watt Univeristy
University of Bradford
Anthony Brabazon
Attila Csenki
Beatriz De La Iglesia
Bruce Graham
Chris Price
Dario Landa-Silva
David Cairns
Fei Chao
Felician Campean
Francisco Chiclana
Gabriel Oltean
George Coghill
George Magoulas
Giovanni Acampora
Girijesh Prasad
Giuseppina Gini
H.K. Lam
Hui Cheng
Jie Zhang
John Bullinaria
John Mccall
Jonathan Fieldsend
Ke Chen
Keeley Crockett
Keshav Dahal
Larry Bull
Longzhi Yang
Marco Wiering
Michalis Mavrovouniotis
Miltos Petridis
Nick Taylor
Plamen Angelov
Rami Qahwaji
Richard Jensen
Shangming Zhou
Shengxiang Yang
Sorin Hintea
Thomas Jansen
Trevor Martin
Vasile Palade
Waleed Al-Nuaimy
University College Dublin
University of Bradford
University of East Anglia
University of Stirling
Aberystwyth University
University of Nottingham
University of Stirling
Xiamen University
University of Bradford
De Montfort University
Technical University of Cluj-Napoca
University of Aberdeen
Birkbeck College
Nottingham Trent University
University of Ulster
Politecnico di Milano
King's College London
Liverpool John Moores University
Newcastle University
University of Birmingham
Robert Gordon University
University of Exeter
The University of Manchester
Manchester Metropolitan University
University of West of Scotland
University of the West of England
Northumbria University
University of Groningen
De Montfort University
University of Brighton
Heriot-Watt University
Lancaster University
University of Bradford
Aberystwyth University
University of Wales Swansea
De Montfort University
Technical University of Cluj-Napoca
Aberystwyth University
University of Bristol
Coventry University
University of Liverpool
8
Wei Pang
Xiao-Jun Zeng
Yaxin Bi
Yingjie Yang
Yonghong Peng
Jinchao Ji
Anna Harutyunyan
Peter Scopes
Haruna Isah
Miqing Li
Vahab Youssofzadeh
Kleanthis Malialis
Pritesh Mistry
Haider Raza
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
University of Aberdeen
University of Manchester
University of Ulster
De Montfort University
University of Bradford
University of Aberdeen
University of York
University of York
University of Bradford
Brunel University
University of Ulster
University of York
University of Bradford
University of Ulster
9
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
UKCI2014 Programme of Events
Monday 8
th
September 2014
0800 – 1200 Registration (Student Central Boardroom SC0.56)
0900 – 0930 Welcome Talk (Student Central Lecture Theatre SC0.51)
Professor Brian Cantor CBE , Vice-Chancellor and Principal of the University of Bradford
0930 – 1030 Invited Keynote Talk
Computational Intelligence Research for Impact in the Games Industry
Professor Peter Cowling , University of York
1030 – 1100 Tea and Coffee Break
1100 – 1240 Session 1: Emerging and Applied Computational Intelligence
Session Chair: Dr Paul Trundle, University of Bradford
Pathfinding in partially explored games environments
John Stamford, Arjab Singh Khuman, Jenny Carter and Samad Ahmadi
Evolution-In-Materio: Solving Function Optimization Problems Using Materials
Maktuba Mohid, Julian F. Miller, Simon L. Harding, Gunnar Tufte, Odd Rune Lykkebø,
Mark K. Massey and Mike C. Petty
Integration Strategies for Toxicity Data From an Empirical Perspective
Longzhi Yang and Daniel Neagu
Nonlinear control of the underactuated hovercraft using the Derivative-free nonlinear
Kalman Filter
Gerasimos Rigatos and Guilherme Raffo
A novel approach for ANFIS modelling based on Grey system theory for thermal error compensation
Ali Mohamed Abdulshahed, Andrew Peter Longstaff and Simon Fletcher
1240 – 1320 Lunch Break
1320 – 1500 Session 2: Advances and Applications in Data Mining
Session Chair: Dr Mariam Kiran, University of Bradford
Integrate Classifier Diversity Evaluation to Feature Selection Based Classifier Ensemble
Reduction
Gang Yao, Fei Chao, Hualin Zeng, Minghui Shi, Min Jiang, Changle Zhou
Data Density Based Clustering
Richard Hyde and Plamen Angelov
Social Media Analysis for Product Safety using Text Mining and Sentiment Analysis
Haruna Isah, Daniel Neagu and Paul Trundle
10
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Kernel Learning Method for Distance-based Classification of Categorical Data
Lifei Chen, Gongde Guo, Shengrui Wang and Xiangzeng Kong
A New Weighting Scheme and Discriminative Approach for Information Retrieval in Static and Dynamic Document Collections
Osman A. S. Ibrahim and Dario Landa-Silva
1500 – 1520 Tea and Coffee Break
1520 – 1700 Session 3: Advances & Applications in Signal and Image Processing
Session Chair: Dr Kenneth McGarry, University of Sunderland
Adaptive Learning with Covariate Shift-Detection for Non-Stationary Environments
Haider Raza, Girijesh Prasad and Yuhua Li
Search-guided Activity Signals Extraction in Application Service Management Control
Tomasz D. Sikora and George D. Magoulas
Modeling Neural Plasticity in Echo State Networks for Time Series Prediction
Mohd-Hanif Yusoff and Yaochu Jin
Neural Networks and Wavelet Transform in Waveform Approximation
Paul Farago, Gabriel Oltean and Laura-Nicoleta Ivanciu
Investigating the Relationship Between the Distribution of Local Semantic Concepts and
Local Keypoints for Image Annotation
Yousef Alqasrawi and Daniel Neagu
1700 – 1730 Guided Walk to Bradford City Hall, Centenary Square, Bradford
1730 – 1830 Lord Mayor: Welcome and Tour of Bradford City Hall
1830: Welcome Reception (Forster’s Bistro and Deli, Centenary Square, Bradford)
11
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Tuesday 9
th
September 2014
0900 – 1000 Invited Keynote Talk
Deriving the most value out of local data
Mark Barrett , Open Data Lead for Leeds
1000 – 1020 Tea and Coffee Break
1020 – 1200 Session 4: Fuzzy Sets and Fuzzy Systems
Session Chair: Dr Longzhi Yang,
Northumbria
University
A Fuzzy Image Congealing-based Handwritten Chinese Character Recognition and
Classification System
Fangyi Li, Qiang Shen, Ying Li and Neil Mac Parthaláin
Fuzzy Adaptive Parameter Control of a Late Acceptance Hyper-heuristic
Warren G. Jackson, Ender Ozcan and Robert I. John
Interval Type-2 Fuzzy Sets in Supplier Selection
Seda T r Ro ert ohn and nder can
Fast Learning method of Interval Type-2 Fuzzy Neural Networks
Damian Olczyk and Urszula Markowska-Kaczmar
Refinement of Fuzzy Rule Weights with Particle Swarm Optimisation
Tianhua Chen, Qiang Shen, Pan Su and Changjing Shang.
1200 – 1330 Lunch Break
1330 - 1510 Session 5: Advances and Applications of Evolutionary Computing and Optimization
Session Chair: Dr Neil Mac Parthaláin, Aberystwyth University
A Benchmark Generator for Dynamic Multi-objective Optimization Problems
Shouyong Jiang and Shengxiang Yang
PermGA Algorithm for a Sequential Optimal Space Filling DoE Framework
Mohammed Reza Kianifar, Felician Campean and Alastair Wood
Hybridisation of Decomposition and GRASP for Combinatorial Multiobjective Optimisation
Ahmad Alhindi, Qingfu Zhang and Edward Tsang
Dynamic Railway Junction Rescheduling using Population Based Ant Colony Optimisation
Jayne Eaton and Shengxiang Yang
Adaptive Mutation in Dynamic Environments
Chigozirim J. Uzor, Mario Gongora, Simon Coupland and Benjamin N. Passow
1510 – 1530 Tea and Coffee Break
12
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
1530 - 1630 Poster Session
Session Chair: Professor Daniel Neagu, University of Bradford
Crowd Formal Modelling and Simulation: The Sa’yee Ritual
Ilias Sakellariou, Omar Kurdi, Marian Gheorghe, Daniela Romano, Petros Kefalas,
Florentin Ipate and Ionut Mihai Niculescu
Soft Morphological Filter Optimization Using a Genetic Algorithm for Noise Elimination
Turker Ercal, Ender Ozcan and Shahriar Asta
A Rule Based System for Diagnosing and Treating Chronic Heart Failure
Luke Vella Critien, Arjab Singh Khuman, Jenny Carter and Samad Ahmadi
Computational Techniques for Identifying Networks of Interrelated Diseases
Ken McGarry and Ukeme Daniel
Word Order Variation and String Similarity Algorithm to Reduce Pattern Scripting in
Pattern Matching Conversational Agents
Mohammed Kaleem, James D. O'Shea and Keeley A. Crockett
Automatic Image Annotation with Long Distance Spatial-Context
Donglin Cao, Dazhen Lin and Jiansong Yu
PCA-based Algorithmic Approximation of Crisp Target Sets
Ray-Ming Chen
Optimized Artificial Neural Network Using Differential Evolution for Prediction of RF Power in VHF/UHF TV and GSM 900 Bands for Cognitive Radio Networks
Sunday Iliya, Eric Goodyer, Mario Gongora, John Gow and Jethro Shell
Truss Topology Optimization with Species Conserving Genetic Algorithm
Jian-Ping Li and Felician Campean
The Effect of Attribute Pairings in Intrusion Detection
Michael Milliken, Yaxin Bi and Leo Galway
Towards Automated Monitoring of Adult Zebrafish
Qussay Al-Jubouri, Waleed Al-Nuaimy, Hamza S. AlZu'bi, Osama Zahran and Jonathan
Buckley
Improving Motion State Change Object Detection by Using Block Background Context
Dazhen Lin, Donglin Cao and Hualin Zeng
1630 - 1730 UKCI Matters: Chair Professor Qiang Shen, Aberystwyth University
UKCI2015 : Professor Richard Everson General Chair, University of Exeter
1830: UKCI2014 Dinner. Best Student Paper Award presentation (Midland Hotel)
13
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Wednesday 10
th
September 2014
0900 - 1000 Session 6: Intelligent Technologies for Healthcare and Behaviour Analysis 1
Session Chair: Professor Rami Qahwaji, University of Bradford
An Efficient System For Preprocessing Confocal Corneal Images For Subsequent Analysis
Mhd Saeed Sharif, Rami Qahwaji, Sofyan Hayajneh, Stan Ipson, Rania Alzubaidi and
Arun Brahma
Human Activity Classification Using A Single Accelerometer
Hamzah S. AlZubi, Simon Gerrard-Longworth, Waleed Al-Nuaimy, Yannis Goulermas and
Stephen Preece
Multiple sensor outputs and computational intelligence towards estimating state and speed for control of lower limb prostheses
Pamela A. Hardaker, Benjamin N. Passow and David Elizondo
1000 – 1020 Tea and Coffee Break
1020 - 1140 Session 7: Intelligent Technologies for Healthcare and Behaviour Analysis 2
Session Chair: Dr Waleed Al-Nuaimy, University of Liverpool
Blood-Glucose Pattern Mining Algorithm for Decision Support in Diabetes Management
Majid A. Al-Taee, Suhail N. Abood, Waleed Al-Nuaimy and Ahmad M. Al-Taee
Integrated Microarray Analytics for the Discovery of Gene Signatures for Triple-Negative
Breast Cancer
Masood U.H. Zaka, Yonghong Peng and Chris W. Sutton
Prediction of Driver Fatigue: Approaches and Open Challenges
Hilal Abbood, Waleed Al-Nuaimy, Ali Al-Ataby, Sameh A. Salem and Hamzah S. AlZubi
Using Computational Methods for the Prediction of Drug Vehicles
Pritesh Mistry, Anna Palczewska, Daniel Neagu and Paul Trundle
1140 – 1230 Lunch Break
1230 – 1330 Invited Keynote Talk
Learning about Activities and Objects from Video
Professor Anthony Cohn , University of Leeds
14
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Keynote Talks
Computational Intelligence Research for Impact in the Games Industry:
Professor Peter Cowling, University of York
Abstract : Digital Games are a huge international industry, with revenues of
£3 billion in the UK and over $70 billion worldwide, far exceeding that of film, music and DVD. Games have been an important testbed for AI/CI research for decades, since pioneering work by Turing and Shannon.
Computational Intelligence is an important component of many (arguably most) commercial digital games. Games provide a superb testbed for CI research, and one whose potential we are only just starting to understand. In this talk we will present significant research advances our team has made in Monte Carlo Tree Search, and discuss how these were used by a commercial games company (AI Factory Ltd.) resulting in our research advances being used by 2.5 million game players worldwide in a charttopping mobile phone game. We will describe how our Monte Carlo Tree Search approaches, tailored for "strong" gameplay, needed to be recast to yield "fun" gameplay as needed by the commercial market. Furthermore, widespread play against our CI produced a deluge of data. Subsequently, we analysed this data to understand how human players react to different CI playing styles - providing new metrics for our CI which would not have been possible without massive amounts of gameplay data from human players. We will round out the talk by discussing other examples where digital games have been used to do science, and by speculating as to the future of games as a tool for research in CI and other fields.
Biography :Peter Cowling wrote his first game playing Articial/Computational Intelligence
(AI), for the board game Othello, at about 9 years of age, on a Commodore PET computer with a mighty 8 kilobytes of RAM (about one tenth of one percent of the size of a single decent digital photograph nowadays :). He was quickly hooked and has created many games and AI for games as a way to test out research ideas in AI during his research career [1], as well as inventing hyperheuristics [2] and extensive work in Operational
Research (see e.g. [3]) with organisations such as AI Systems BV, Gaist Solutions Ltd,
Trimble MRM and the Nationwide Building Society. Now, with the rise and rise of the games industry, games is an economically and socially important area of research and is a principal area of his work. He currently leads the EPSRC Centre for Doctoral Training in
Intelligent Games and Game Intelligence (IGGI www.iggi.org.uk
) and the EPSRC/ESRC
New Economic Models and Opportunities for Digital Games (NEMOG) projects
( www.nemog.org
), as well as collaborative projects in optimisation and decision-making.
[1] Edward J. Powley, Peter I. Cowling, Daniel Whitehouse (2014) "Information capture and reuse strategies in Monte Carlo Tree Search, with applications to games of hidden information", Artificial Intelligence, to appear. DOI:10.1016/j.artint.2014.08.002.
[2] P. Cowling, G. Kendall, E. Soubeiga (2001) "A Hyperheuristic Approach to Scheduling a Sales Summit", in selected papers from 3rd International Conference on the Practice and Theory of Automated Timetabling (PATAT), Konstanz, Germany, Springer LNCS
2079, 176-190.
[3] S.M. Remde, P.I. Cowling, K.P. Dahal, N.J. Colledge, E. Selensky (2011) "An empirical study of hyperheuristics for managing very large sets of low level heuristics", Journal of the
Operational Research Society, 63 (3): 392-405.
15
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Deriving the most value out of local data:
Mark Barrett, Open Data Lead for Leeds
Abstract : Data.gov.uk
is the central repository for Open Data in the UK and holds nearly 20,000 datasets. Of these 20,000 only 3% come from local councils and are classified as "Open". With such a disparity between nationally available data and locally available data, there is a lot of work needed to close the gap. Significant amounts of data are held behind closed doors, but could be made "Open". This provides opportunities for cities to release data, from across sectors, so that citizens can understand their cities in new exciting ways. Over the last 9 months we have worked to release increasing amounts of data, engage with citizens and create a new economy in the city. This talk will give highlight the work that's been carried out.
Biography : Mark Barrett is an Open Data Lead for Leeds. He created Leeds Data Mil l - the Open Data platform for the city that holds public sector, private sector, and third sector data so create a rich picture of Leeds. Mark co-founded Leeds Data Thing , and created the 1st Open Data app to reach #1 in iTunes GP Ratings and was selected as one of the top 50 innovators in healthcare by the Health Service Journal in 2013. Marks passion is for local level data - helping people to understand their environment and improve their surroundings.
Learning about Activities and Objects from Video:
Professor Anthony Cohn, University of Leeds
Abstract : In this talk I will present ongoing work at Leeds on building models of video activity. I will present techniques, both supervised and unsupervised, for learning the spatio-temporal structure of tasks and events from video or other sensor data, particularly in the case of scenarios with concurrent activities. In both cases, the representation will exploit qualititive spatio-temporal relations. A novel method for robustly transforming video data to qualitative relations will be presented. I will also show how objects can be "functionally categorised" according to their spatio-temporal behaviour.
Biography : Tony Cohn holds a Personal Chair at the University of Leeds, where he is
Professor of Automated Reasoning. He is presently Director of the Institute for Artificial
Intelligence and Biological Systems. His work on Knowledge Representation and
Reasoning has a particular focus on qualitative spatial/spatio-temporal reasoning, the best known being the well cited Region Connection Calculus (RCC). His current research interests range from theoretical work on spatial calculi and spatial ontologies, to cognitive vision, modelling spatial information in the hippocampus, and detecting buried underground assets (e.g. utilities and archaeological residues) using a variety of geolocated sensors. He has been Chairman/President of SSAISB, ECCAI, KR inc, the IJCAI
Board of Trustees and is presently Editor-in-Chief of the AAAI Press, Spatial Cognition and
Computation, and the Artificial Intelligence journal. He was elected a founding Fellow of
ECCAI, and is also a Fellow of AAAI, AISB, the BCS, and the IET. Work from the Cogvis project won the British Computer Society Machine Intelligence prize in 2004, and the
VAULT system from his Mapping the Underworld project won a 2012 IET Innovation
Award.
16
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Papers
Session 1: Emerging and Applied Computational Intelligence
Pathfinding in partially explored games environments. The application of the A* Algorithm with
Occupancy Grids in Unity3D
John Stamford, Arjab Singh Khuman, Jenny Carter & Samad Ahmadi
Centre for Computational Intelligence
School of Computing
De Montfort University
Leicester, United Kingdom
Abstract —One of the key aspects of games development is making Non-Playable
Characters (NPC) behave more realistically in the environment. One of the main challenges is creating an NPC that is aware of its surroundings and acts accordingly. The
A* Algorithm is widely used in the games development community to allow AI based characters to move around the environment however unlike real characters they are often given information about the environment without having to explore. This paper combines the use of the A* Algorithm with the occupancy grid technique to allow Non-Playable
Characters to build their own representation of the environment and plan paths based on this information. The paper demonstrates the application of the approach and shows a range of testing and its limitations.
Evolution-In-Materio: Solving Function Optimization Problems Using Materials
Maktuba Mohid1, Julian F. Miller1, Simon L. Harding1, Gunnar Tufte2, Odd Rune
Lykkebø2, Mark K. Massey3, and Michael C. Petty3
1Department of Electronics, University of York, York, UK.
2Department of Computer and Information Science, Norwegian University of Science and
Technology, 7491
Trondheim, Norway.
3School of Engineering and Computing Sciences and Centre for Molecular and Nanoscale
Electronics, Durham University, UK.
Abstract
—Evolution-in-materio (EIM) is a method that uses artificial evolution to exploit properties of materials to solve computational problems without requiring a detailed understanding of such properties. In this paper, we show that using a purpose-built hardware platform called Mecobo, it is possible to evolve voltages and signals applied to physical materials to solve computational problems. We demonstrate for the first time that this methodology can be applied to function optimization. We evaluate the approach on 23 function optimization benchmarks and in some cases results come very close to the global optimum or even surpass those provided by a well-known software-based evolutionary approach. This indicates that EIM has promise and further investigations would be fruitful.
Integration Strategies for Toxicity Data from an Empirical Perspective
Longzhi Yang
Department of Computer Science and Digital Technologies, Northumbria University
Daniel Neagu
Department of Computing, School of Electrical Engineering and Computer Science
University of Bradford
Abstract —The recent development of information techniques, especially the state-of-theart “ ig data” solutions ena les the extracting, gathering, and processing large amount of toxicity information from multiple sources. Facilitated by this technology advance, a
17
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) framework named integrated testing strategies (ITS) has been proposed in the predictive toxicology domain, in an effort to intelligently jointly use multiple heterogeneous toxicity data records (through data fusion, grouping, interpolation/extrapolation etc.) for toxicity assessment. This will ultimately contribute to accelerating the development cycle of chemical products, reducing animal use, and decreasing development costs. Most of the current study in ITS is based on a group of consensus processes, termed weight of evidence (WoE), which quantitatively integrate all the relevant data instances towards the same endpoint into an integrated decision supported by data quality. Several WoE implementations for the particular case of toxicity data fusion have been presented in the literature, which are collectively studied in this paper. Noting that these uncertainty handling methodologies are usually not simply developed from conventional probability theory due to the unavailability of big datasets, this paper first investigates the mathematical foundations of these approaches. Then, the investigated data integration models are applied to a representative case in the predictive toxicology domain, with the experimental results compared and analysed.
Nonlinear control of the underactuated hovercraft using the Derivative-free nonlinear Kalman
Filter
Gerasimos Rigatos
Unit of Industrial Automation, Industrial Systems Institute 26504, Rion Patras, Greece
Guilherme Raffo
Department of Electronic Engineering, Federal University of Minas Gerais, 31270-901 -
Belo Horizonte, Brazil
Abstract
—The paper proposes a nonlinear control approach for the underactuated hovercraft model based on differential flatness theory and using a new nonlinear state vector and disturbances estimation method under the name of Derivativefree nonlinear
Kalman Filter. It is proven that the sixth order nonlinear model of the hovercraft is a differentially flat one. It is shown that this model cannot be subjected to static feedback linearization, however it admits dynamic feedback linearization which means that the system’s state vector is extended y including as additional state variables the control inputs and their derivatives. Next, using the differential flatness properties it is also proven that this model can be subjected to input-output linearization and can be transformed to an equivalent canonical (Brunovsky) form. Based on this latter description the design of a state feedback controller is carried out enabling accurate maneuvering and trajectory tracking. Additional problems that are solved in the design of this feedback control scheme are the estimation of the nonmeasura le state varia les in the hovercraft’s model and the compensation of modeling uncertainties and external perturbations affecting vessel. To this end, the application of the Derivative-free nonlinear Kalman Filter is proposed. This nonlinear filter consists of the Kalman Filter’s recursion on the linearized equivalent of the vessel and of an inverse nonlinear transformation based on the differential flatness features of the system which enables to compute state estimates for the state variables of the initial nonlinear model. The redesign of the filter as a disturbance observer makes possible the estimation and compensation of additive perturbation terms affecting the hovercraft’s model. The efficiency of the proposed nonlinear control and state estimation scheme is confirmed through simulation experiments.
A novel approach for ANFIS modelling based on Grey system theory for thermal error compensation
Ali M Abdulshahed, Andrew P Longstaff and Simon Fletcher
Centre for Precision Technologies, University of Huddersfield, Huddersfield, UK
18
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Abstract
— The fast and accurate modelling of thermal errors in machining is an important aspect for the implementation of thermal error compensation. This paper presents a novel modelling approach for thermal error compensation on CNC machine tools. The method combines the Adaptive Neuro Fuzzy Inference System (ANFIS) and Grey system theory to predict thermal errors in machining. Instead of following a traditional approach, which utilises original data patterns to construct the ANFIS model, this paper proposes to exploit
Accumulation Generation Operation (AGO) to simplify the modelling procedures. AGO, a basis of the Grey system theory, is used to uncover a development tendency so that the features and laws of integration hidden in the chaotic raw data can be sufficiently revealed.
AGO properties make it easier for the proposed model to design and predict. According to the simulation results, the proposed model demonstrates stronger prediction power than standard ANFIS model only with minimum number of training samples.
Session 2: Advances and Applications in Data Mining
Integrate Classifier Diversity Evaluation to Feature Selection Based Classifier Ensemble
Reduction
Gang Yao, Fei Chao, Hualin Zeng, Minghui Shi, Min Jiang, Changle Zhou
Cognitive Science Department, School of Information Science and Engineering, Fujian
Provincial Key Lab of Brain-like Intelligent Systems, Xiamen University, China 361005
Abstract
—Classifier ensembles improve the performance of single classifier system.
However, a classifier ensemble with too many classifiers may occupy a large number of computational time. This paper proposes a new ensemble subset evaluation method that integrates classifier diversity measures into a classifier ensemble reduction framework.
The approach is implemented by using three conventional diversity algorithms and one new developed diversity measure method to calculate the diversity’s merits within the classifier ensemble reduction framework. The subset evaluation method is demonstrated by the experimental data: the method not only can meet the requirements of high accuracy rate and fewer size, but also its running time is greatly shortened. When the accuracy requirements are not very strict, but the running time requirements is more stringent, the proposed method is a good choice.
Data Density Based Clustering
Richard Hyde, Plamen Angelov
Data Science Group, School of Computing and Communications, Lancaster University
Lancaster, LA1 4WA, UK
Abstract
— A new, data density based approach to clustering is presented which automatically determines the number of clusters. By using RDE for each data sample the number of calculations is significantly reduced in offline mode and, further, the method is suitable for online use. The clusters allow a different diameter per feature/dimension creating hyper-ellipsoid clusters which are axisorthogonal. This results in a greater differentiation between clusters where the clusters are highly asymmetrical.
We illustrate this with 3 standard data sets, 1 artificial dataset and a large real dataset to demonstrate comparable results to Subtractive, Hierarchical, K-Means, ELM and DBScan clustering techniques. Unlike subtractive clustering we do not iteratively calculate P however. Unlike hierarchical we do not need O(N2) distances to be calculated and a cutoff threshold to be defined. Unlike k-means we do not need to predefine the number of clusters.
Using the RDE equations to calculate the densities the algorithm is efficient, and requires no iteration to approach the optimal result. We compare the proposed algorithm to k-
19
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) means, subtractive, hierarchical, ELM and DBScan clustering with respect to several criteria. The results demonstrate the validity of the proposed approach.
Social Media Analysis for Product Safety using Text Mining and Sentiment Analysis
Haruna Isah, Paul Trundle, Daniel Neagu
Artificial Intelligence Research (AIRe) Group, School of Electrical Engineering and
Computer Science, University of Bradford, Bradford, UK
Abstract — The growing incidents of counterfeiting and associated economic and health consequences necessitate the development of active surveillance systems capable of producing timely and reliable information for all stake holders in the anticounterfeiting fight.
User generated content from social media platforms can provide early clues about product allergies, adverse events and product counterfeiting. This paper reports a work in progress with contributions including: the development of a framework for gathering and analyzing the views and experiences of users of drug and cosmetic products using machine learning, text mining and sentiment analysis; the application of the proposed framework on
Facebook comments and data from Twitter for brand analysis, and the description of how to develop a product safety lexicon and training data for modeling a machine learning classifier for drug and cosmetic product sentiment prediction. The initial brand and product comparison results signify the usefulness of text mining and sentiment analysis on social media data while the use of machine learning classifier for predicting the sentiment orientation provides a useful tool for users, product manufacturers, regulatory and enforcement agencies to monitor brand or product sentiment trends in order to act in the event of sudden or significant rise in negative sentiment.
Kernel Learning Method for Distance-Based Classification of Categorical Data
Lifei Chen, Gongde Guo
School of Mathematics and Computer Science, Fujian Normal University, Fujian 350117,
China
Shengrui Wang
Department of Computer Science, University of Sherbrooke, Quebec J1K 2R1, Canada
Xiangzeng Kong
School of Computing and Mathematics, University of Ulster at Jordanstown, Northern
Ireland BT37 0QB, UK
Abstract —Kernel-based methods have become popular in machine learning; however, they are typically designed for numeric data. These methods are established in vector spaces, which are undefined for categorical data. In this paper, we propose a new kind of kernel trick, showing that mapping of categorical samples into kernel spaces can be alternatively described as assigning a kernel-based weight to each categorical attribute of the input space, so that common distance measures can be employed. A data-driven approach is then proposed to kernel bandwidth selection by optimizing feature weights.
We also make use of the kernel-based distance measure to effectively extend nearestneighbor classification to classify categorical data. Experimental results on realworld data sets show the outstanding performance of this approach compared to that obtained in the original input space.
A New Weighting Scheme and Discriminative Approach for Information Retrieval in Static and
Dynamic Document Collections
Osman A. S. Ibrahim
School of Computer Science, ASAP Research Group, The University of Nottingham, UK
Computer Science Dept., Minia University, Egypt
20
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Dario Landa-Silva
School of Computer Science, ASAP Research Group, The University of Nottingham, UK
Abstract —This paper introduces a new weighting scheme in information retrieval. It also proposes using the document centroid as a threshold for normalizing documents in a document collection. Document centroid normalization helps to achieve more effective information retrieval as it enables good discrimination between documents. In the context of a machine learning application, namely unsupervised document indexing and retrieval, we compared the effectiveness of the proposed weighting scheme to the ‘Term Frequency
– Inverse Document Frequency’ or TF-IDF, which is commonly used and considered as one of the best existing weighting schemes. The paper shows how the document centroid is used to remove less significant weights from documents and how this helps to achieve better retrieval effectiveness. Most of the existing weighting schemes in information retrieval research assume that the whole document collection is static. The results presented in this paper show that the proposed weighting scheme can produce higher retrieval effectiveness compared with the TF-IDF weighting scheme, in both static and dynamic document collections. The results also show the variation in information retrieval effectiveness that is achieved for static and dynamic document collections by using a specific weighting scheme. This type of comparison has not been presented in the literature before.
Session 3: Advances and Applications in Signal and Image Processing
Adaptive Learning with Covariate Shift-Detection for Non-Stationary Environments
Haider Raza, Girijesh Prasad, Yuhua Li
Intelligent Systems Research Center, University of Ulster, Londonderry, Northern Ireland,
UK
Abstract
—Learning with dataset shift is a major challenge in non-stationary environments wherein the input data distribution may shift over time. Detecting the dataset shift point in the time-series data, where the distribution of time-series shifts its properties, is of utmost interest. Dataset shift exists in a broad range of real-world systems. In such systems, there is a need for continuous monitoring of the process behavior and tracking the state of the shift so as to decide about initiating adaptation in a timely manner. This paper presents an adaptive learning algorithm with dataset shift-detection using an exponential weighted moving average (EWMA) model based test in a non-stationary environment. The proposed method initiates the adaptation by reconfiguring the knowledge-base of the classifier. This algorithm is suitable for real-time learning in non-stationary environments. Its performance is evaluated through experiments using synthetic datasets. Results show that it reacts well to different covariate shifts.
Search-guided Activity Signals Extraction in Application Service Management Control
Tomasz D. Sikora and George D. Magoulas
Department of Computer Science and Information Systems, Birkbeck, University of
London, Malet Street, London WC1E 7HX
Abstract —The increased interest in autonomous control in Application Service
Management-ASM environments has driven the demand for analysis of multivariate datasets in this area. Gathered metrics form time-series that can be considered as signals, which should be decomposed in order to find relations between system utilization and effective activity. This paper introduces a metrics signal deconvolution method that can be used to support human administrators or can be incorporated into feature extraction schemes that feed decision blocks of autonomous controllers. The method considers ASM
21
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) environments signals decomposition as a search problem that is solved using heuristics and metaheuristic strategies. Quantitative and qualitative relations between activity and system resources signals are searched with use of a model that is based on similarity and variability of the changes, under minimal assumptions about the ASM system architecture and design. Experimental results show that the model can be successfully integrated with optimization techniques and the results produced when tested using data produced through queue modeling meet human perception of the signal unmixing problem.
Modeling Neural Plasticity in Echo State Networks for Time Series Prediction
Mohd-Hanif Yusoff and Yaochu Jin
Department of Computing, FEPS, University of Surrey, Guildford, United Kingdom
Abstract — In this paper, we investigate the influence of neural plasticity on the learning performance of echo state networks (ESNs) and supervised learning algorithms in training readout connections for two time series prediction problems including the sunspot time series and the Mackey Glass chaotic system. We implement two different plasticity rules that are expected to improve the prediction performance, namely, anti-Oja learning rule and the Bienenstock-Cooper-Munro (BCM) learning rule combined with both offline and online learning of the read-out connections. Our experimental results have demonstrated that the neural plasticity can more significantly enhance the learning in offline learning than in online learning.
Neural Networks and Wavelet Transform in Waveform Approximation
Paul Farago, Gabriel Oltean, Laura-Nicoleta Ivanciu
Bases of Electronics Department, Technical University of Cluj-Napoca, Romania
Abstract
—To fully analyze the time response of a complex system, in order to discover its critical operation points, the output waveform (under all conceivable conditions) needs to be generated. Using conventional methods as physical experiments or detailed simulations can be prohibitive from the resources point of view (time, equipment). The challenge is to generate the waveform by its numerous time samples as a function of different operating conditions described by a set of parameters. In this paper, we propose a fast to evaluate, but also accurate model that approximates the waveforms, as a reliable substitute for complex physical experiments or overwhelming system simulations. Our proposed model consists of two stages. In the first stage, a previously trained artificial neural network produces some coefficients standing for “primary” coefficients of a wavelet transform. In the second stage, an inverse wavelet transform generates all the time samples of the expected waveform using a fusion etween the “primary” coefficients and some “secondary” coefficients previously extracted from the nominal waveform in the family. The test results for a number of 100 different combinations of three waveform parameters show that our model is a reliable one, featuring high accuracy and generalization capabilities, as well as high computation speed.
Investigating the relationship between the distribution of local semantic concepts and local keypoints for image annotation
Yousef Alqasrawi
Department of Computer Science, Applied Science University, Amman, Jordan
Daniel Neagu
Department of Computing, University of Bradford, UK
Abstract
— The problem of image annotation has gained increasing attention from many researchers in computer vision. Few works have addressed the use of bag of visual words for scene annotation at region level. The aim of this paper is to study the relationship
22
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) between the distribution of local semantic concepts and local keypoints located in image regions labelled with these semantic concepts. Based on this study, we investigate whether bag of visual words model can be used to efficiently represent the content of natural scene image regions, so images can be annotated with local semantic concepts.
Also, this paper presents local from global approach which study the influence of using visual vocabularies generated from general scene categories to build bag of visual words at region level. Extensive experiments are conducted over a natural scene dataset with six categories. The reported results have shown the plausibility of using the BOW model to represent the semantic information of image regions.
Session 4: Fuzzy Sets and Fuzzy Systems
A Fuzzy Image Congealing-based Handwritten Chinese Character Recognition and Classification
System
Fangyi Li
1
, Qiang Shen
2
, Ying Li
1
and Neil Mac Parthaláin
2
1
School of Computer Science, Northwestern Polytechnical University, Xian 710072, China
2
Department of Computer Science, Institute of Mathematics, Physics and Computer
Science (IMPACS), Aberystwyth University, UK
Abstract —The recognition and classification of handwritten Chinese characters poses a significant challenge for automated methods. Indeed the sheer number of characters, intricate complexity of such characters, and variations in writing styles mean that the task can be difficult even for humans. Previous work in this area has focused upon methods which perform a certain form of feature extraction and segmentation as the basis for building systems to perform this task. This paper proposes two approaches for handwritten
Chinese character recognition and classification using an image alignment technique based on a fuzzy-entropy metric. Rather than extracting features from the image, which can often result in subjective and poorly-fitting models, the proposed methods instead uses the mean image transformations of the training phase as a basis for building models.
The use of a fuzzy-entropy based metric also means improved ability to model different types of uncertainty. The mean image transformations are then collated, and used as training data to classify the images of test characters. A nearest-neighbour classifier based on Euclidean distance is then used to classify each test character. The approaches are applied to a publicly available real-world database of handwritten Chinese characters and demonstrate that they can achieve high classification accuracy.
Fuzzy Adaptive Parameter Control of a Late Acceptance Hyper-heuristic
Warren G. Jackson, Ender Ozcan and Robert I. John
ASAP Research Group, School of Computer Science, University of Nottingham
Abstract —A traditional iterative selection hyper-heuristic which manages a set of low level heuristics relies on two core components, a method for selecting a heuristic to apply at a given point, and a method to decide whether or not to accept the result of the heuristic application. In this paper, we present an initial study of a fuzzy system to control the listsize parameter of late acceptance move acceptance method as a selection hyper-heuristic component. The performance of the fuzzy controlled selection hyper-heuristic is compared to its fixed parameter version and the best hyper-heuristic from a competition on the MAX-
SAT problem domain. The results illustrate that a fuzzy control system can potentially be effective within a hyper-heuristic improving its performance.
Interval Type-2 Fuzzy Sets in Supplier Selection
Seda T r Ro ert ohn and nder an
23
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
ASAP Research Group, School of Computer Science, University of Nottingham, UK
Abstract —Selection of an appropriate supplier is a crucial and challenging task in the effective management of a supply chain. This study introduces a model for solving the supplier selection problem using interval type-2 fuzzy sets. Moreover, the influence of the membership function shape on the results obtained from the model has been investigated on a real-world problem instance tackled by Ordoobadi.
Fast Learning method of Interval Type-2 Fuzzy Neural Networks
Damian Olczyk and Urszula Markowska-Kaczmar
Wroclaw University of Technology Wyb. Wyspianskiego 27, 50-370 Wroclaw, Poland
Abstract —The Fuzzy Set Parameter Estimation algorithm is proposed for fast learning interval type-2 fuzzy neural networks applied for classification problems. Classes are disjoint. Learning consists of estimating appropriate values of fuzzy set parameters in every rule. Estimation is based on statistical properties of the training data. The experimental study confirms that it is dozens times quicker than the backpropagation method, while the classification effectiveness is comparable.
Refinement of Fuzzy Rule Weights with Particle Swarm Optimisation
Tianhua Chen, Qiang Shen, Pan Su and Changjing Shang.
Department of Computer Science, Institute of Mathematics, Physics and Computer
Science, Aberystwyth University, UK
Abstract —The most challenging problem in the design of fuzzy rule-based classification systems is the construction of a fuzzy rule base for the target problem. Much research has focused on generating and adjusting antecedent fuzzy sets. In many cases, initial fuzzy sets, each of which has a linguistic meaning, are predefined by domain experts and are thus required to be maintained in order to ensure interpretability of any subsequent inference results. However, learning fuzzy rules using fixed fuzzy quantity space without any quantification will restrict the accuracy of the resulting rules. Fortunately, adjusting the weight of a fuzzy if-then rule may help improve classification accuracy without degrading the interpretability. There have been different proposals for fuzzy rule weight tuning through the use of various heuristics with limited success. This paper proposes an alternative approach using Particle Swarm Optimisation in the search of a set of optimal rule weights, which can entail high classification accuracy. The proposed method is initially tested on the iris data set with regard to different predefined fuzzy partitions of linguistic variables to assess its performance. Experimental results demonstrate that the proposed approach is not sensitive to the predefined fuzzy partitions, and can boost classification performance especially when a coarse fuzzy partition is given.
Session 5: Advances and Applications of Evolutionary Computing and Optimization
A Benchmark Generator for Dynamic Multi-objective Optimization Problems
Shouyong Jiang and Shengxiang Yang
Centre for Computational Intelligence (CCI), School of Computer Science and Informatics,
De Montfort University, Leicester LE1 9BH, UK
Abstract —Many real-world optimization problems appear to not only have multiple objectives that conflict each other but also change over time. They are dynamic multiobjective optimization problems (DMOPs) and the corresponding field is called dynamic multi-objective optimization (DMO), which has gained growing attention in recent years.
However, one main issue in the field of DMO is that there is no standard test suite to determine whether an algorithm is capable of solving them. This paper presents a new
24
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) benchmark generator for DMOPs that can generate several complicated characteristics, including mixed Pareto-optimal front (convexity-concavity), strong dependencies between variables, and a mixed type of change, which are rarely tested in the literature.
Experiments are conducted to compare the performance of five state-of-the-art DMO algorithms on several typical test functions derived from the proposed generator, which gives a better understanding of the strengths and weaknesses of these tested algorithms for DMOPs.
PermGA Algorithm for a Sequential Optimal Space Filling DoE Framework
Mohammed Reza Kianifar, Felician Campean and Alastair Wood
School of Engineering, University of Bradford, Bradford, United Kingdom
Abstract — This paper presents the development and implementation of a customised
Permutation Genetic Algorithm (PermGA) for a sequential Design of Experiment (DoE) framework based on space filling Optimal Latin Hypercube (OLH) designs. The work is motivated by multivariate engineering problems such as engine mapping experiments, which require efficient DoE strategies to minimise expensive testing. The DoE strategy is based on a flexible Model Building – Model Validation (MB-MV) sequence based on space filling OLH DoEs, which preserves the space filling and projection properties of the DoEs through the iterations. A PermGA algorithm was developed to generate MB OLHs, subsequently adapted for generation of infill MV test points as OLH DoEs, preserving good space filling and projection properties for the merged MB + MV test plan. The algorithm was further modified to address issues with non-orthogonal design spaces. A case study addressing the steady state engine mapping of a Gasoline Direct Injection was used to illustrate and validate the practical application of MBMV sequence based on the developed
PermGA algorithm.
Hybridisation of Decomposition and GRASP for Combinatorial Multiobjective Optimisation
Ahmad Alhindi, Qingfu Zhang and Edward Tsang
School of Computer Science and Electronic Engineering, University of Essex, Colchester,
United Kingdom
Abstract
—This paper proposes an idea of using heuristic local search procedures specific for single-objective optimisation in multiobjectie evolutionary algorithms (MOEAs). In this paper, a multiobjective evolutionary algorithm based on decomposition (MOEA/D) hybridised with a multi-start single-objective metaheuristic called greedy randomised adaptive search procedure (GRASP). In our method a multiobjetive optimisation problem
(MOP) is decomposed into a number of single-objecive subproblems and optimised in parallel by using neighbourhood information. The proposed GRASP alternates between subproblems to help them escape local Pareto optimal solutions. Experimental results have demonstrated that MOEA/D with GRASP outperforms the classical MOEA/D algorithm on the multiobjective 0-1 knapsack problem that is commonly used in the literature. It has also demonstrated that the use of greedy genetic crossover can significantly improve the algorithm performance.
Dynamic Railway Junction Rescheduling using Population Based Ant Colony Optimisation
Jayne Eaton and Shengxiang Yang
Centre for Computational Intelligence (CCI), School of Computer Science and Informatics,
De Montfort University, The Gateway, Leicester LE1 9BH, UK
Abstract —Efficient rescheduling after a perturbation is an important concern of the railway industry. Extreme delays can result in large fines for the train company as well as dissatisfied customers. The problem is exacerbated by the fact that it is a dynamic one;
25
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) more timetabled trains may be arriving as the perturbed trains are waiting to be rescheduled. The new trains may have different priorities to the existing trains and thus the rescheduling problem is a dynamic one that changes over time.
The aim of this research is to apply a population-based ant colony optimisation algorithm to address this dynamic railway junction rescheduling problem using a simulator modelled on a real-world junction in the UK railway network. The results are promising: the algorithm performs well, particularly when the dynamic changes are of a high magnitude and frequency.
Adaptive Mutation in Dynamic Environments
Chigozirim J. Uzor, Mario Gongora, Simon Coupland
Centre for Computational Intelligence (CCI), De Montfort University, Leicester, United
Kingdom
Benjamin N. Passow
DIGITS and CCI, De Montfort University, Leicester, United Kingdom
Abstract
—The interest in nature inspired optimization in dynamic environments has been increasing constantly over the past years. This trend implies that many real world problems experience dynamic changes and it is important to develop an efficient algorithm capable of tackling these problems. Several techniques have been developed over the past two decades for solving dynamic optimization problems. Among these techniques, the hypermutation scheme has proved to be beneficial in solving some of the dynamic optimization problems but requires that the mutation factors be picked a priori.
This paper investigates a new mutation and change detection scheme for compact genetic algorithm (cGA), where the degree of change regulates the mutation rate (i.e. mutation rate is directly proportional to the degree of change). The experimental results shows that the mutation and change detection scheme has an impact on the performance of the cGA in dynamic environments and that the effect of the proposed scheme depends on the dynamics of the environment.
Poster Session
Crowd Formal Modelling and Simulation: The Sa’yee Ritual
Ilias Sakellariou
∗ Omar Kurdi† Marian Gheorghe† Daniela Romano† Petros Kefalas‡
Florentin Ipate°◊ and Ionut Ni ules u°◊
∗
Department of Applied Informatics, University of Macedonia, Thessaloniki, Greece
† Department of Computer Science University of Sheffield, Sheffield, UK
‡ Department of Computer Science The University of Sheffield International Faculty
CITY College, Thessaloniki, Greece
°◊ Department of Computer Science University of Bucharest Bucharest Romania
Abstract —There is an increasing interest in modelling of agents interacting as crowd and a simulation of such scenarios that map to real-life situations. This paper presents a generic state-based abstract model for crowd behaviour that can be mapped onto different agentbased systems. In particular, the abstract model is mapped into the simulation framework
NetLogo. We have used the model to simulate a real-life case study of high density diverse crowd such as the Hajj ritual at the mosque in Mecca (Makkah). The computational model is based on real data extracted from videos of the ritual. We also present a methodology for extracting significant data, parameters, and patterns of behaviour from real-world videos that has been used as an early stage validation to demonstrate that the obtained simulations are realistic.
26
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Soft Morphological Filter Optimization Using a Genetic Algorithm for Noise Elimination
Turker Ercal
Yeditepe University, Department of Computer Engineering Kayıs¸da˘gı 34755 Istan ul,
Turkey
Ender Ozcan and Shahriar Asta
School of Computer Science, University of Nottingham, Nottingham NG8 1BB, UK
Abstract —Digital image quality is of importance in almost all image processing applications. Many different approaches have been proposed for restoring the image quality depending on the nature of the degradation. One of the most common problems that cause such degradation is impulse noise. In general, well known median filters are preferred for eliminating different types of noise. Soft morphological filters are recently introduced and have been in use for many purposes. In this study, we present a Genetic
Algorithm (GA) which combines different objectives as a weighted sum under a single evaluation function and generates a soft morphological filter to deal with impulse noise, after a training process with small images. The automatically generated filter performs better than the median filter and achieves comparable results to the best known filters from the literature over a set of benchmark instances that are larger than the training instances.
Moreover, although the training process involves only impulse noise added images, the same evolved filter performs better than the median filter for eliminating Gaussian noise as well.
A Rule Based System for Diagnosing and Treating Chronic Heart Failure
Luke Vella Critien, Arjab Singh Khuman, Jenny Carter and Samad Ahmadi
Centre for Computational Intelligence, De Montfort University, Leicester, UK
Abstract — The aim of this study is to design a rule based expert system which provides doctors with a better tool to be able to manage chronic heart failure in adults. The suggested system is intended to help in diagnosing heart failure at the earliest permissible stage and subsequently suggest the best treatment for the particular case. The designed system has two facets, one related to the diagnosis and the other related to the treatment of chronic heart failure. The former part is based on the latest chronic heart failure guidelines issued by the National Health Services (NHS) - National Institute for Health and
Clinical Excellence (NICE) in August 2010. The treatment for chronic heart failure is based on the latest version of British National Formulary (BNF). This rule based system is not intended to replace the specialist but it may be used to provide assurance that all diagnostic criteria have been followed and hence the best possible treatment is given. The system is implemented using the CLIPS language which is a powerful forward-chaining rule based system.
Computational Techniques for Identifying Networks of Interrelated Diseases
Ken McGarry and Ukeme Daniel
Department of Pharmacy, Health and wellbeing, Faculty of Applied Sciences, University of
Sunderland, UK
Abstract —Recently there has been a lot of interest in using computational techniques to build networks of protein-to-protein interactions, interacting gene networks and metabolic reactions. Many interesting and novel discoveries have been made using graph based structures using links and nodes to represent the relationships between proteins and genes. Analysis of graph networks has revealed that genes and proteins cooperate in modules performing specific functions and that there is crosstalk or overlap between modules. In this paper we take these ideas further and build upon current knowledge to build up a network of human related diseases based on graph theory and the concept
27
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) of overlap or shared function. We explore the hypothesis that many human diseases are linked by common genetic modules, therefore a defect in one of any of the cooperating genes in a module may lead to a specific disease or related symptom. We build our networks using data and information extracted from several online databases along with supporting knowledge in the form of biological ontologies.
Word Order Variation and String Similarity Algorithm to Reduce Pattern Scripting in Pattern
Matching Conversational Agents
Mohammed Kaleem, James D. O'Shea and Keeley A. Crockett
Department of Computing, Mathematics and Digital Technology, Manchester Metropolitan
University, Manchester, United Kingdom
Abstract — This paper presents a novel sentence similarity algorithm designed to mitigate the issue of free word order in the Urdu language. Free word order in a language poses many challenges when implemented in a conversational agent, primarily due to the fact that it increases the amount of scripting time needed to script the domain knowledge. A language with free word order like Urdu means a single phrase/utterance can be expressed in many different ways using the same words and still be grammatically correct.
This led to the research of a novel string similarity algorithm which was utilized in the development of an Urdu conversational agent. The algorithm was tested through a black box testing methodology which involved processing different variations of scripted patterns through the system to gauge the performance and accuracy of the algorithm with regards to recognizing word order variations of the related scripted patterns. Initial testing has highlighted that the algorithm is able to recognize legal word order variations and reduce the knowledge base scripting of conversational agents significantly. Thus saving great time and effort when scripting the knowledge base of a conversational agent.
Automatic Image Annotation with Long Distance Spatial-Context
Donglin Cao, Dazhen Lin and Jiansong Yu
Cognitive Science Department, ShenZhen Research Institute of Xiamen University, Fujian
Key Laboratory of the Brain-like Intelligent Systems, Xiamen University, Xiamen, China
Abstract —Motion state change object detection, such as stopped objects detection, is one of important topics in Video Surveillance Systems. Generally, backgrounds in the most
Video Surveillance Systems have the property of pureness and selfsimilarity. In this paper, we propose a block background context based background model to solve the motion state change problem. Unlike the classical background model, our approach first models blocks of background, and then determines the learning rate of each block background model by using the block background context information. There are two main advantages. First, the model adaptively selects the learning rate for each block of background model, and that is more flexible than the adaptive learning rate for the whole background. Second, context information helps the determination of true foreground and brings in more reliable information in foreground detection. Our experiments results show that our model outperforms the higher and lower learning rate Gaussian mixture background model in motion state change object detection.
PCA-based Algorithmic Approximation of Crisp Target Sets
Ray-Ming Chen
Department of Computer Science, Friedrich-Alexander-Universitat Erlangen-Nurnberg,
Martensstraße 3, 91058 Erlangen, Germany
28
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Abstract
—Principal Component Analysis (PCA) is an important technique in finding uncorrelated variables. It is applied in many fields: machine learning, pattern recognition, data mining, compression,...,etc. In this paper, we introduce this technique into approximation reasoning. Before the introduction, we construct a theoretical framework of such approximation first. This approximation is based on reasoning of incomplete information in which there exists no algorithm such that the intersection between arbitrary target sets and partitioned clusters is decidable, while there exist some algorithms for the decidability of the subset operation between them. Then, under this framework, we utilize
PCA to implement such approximation reasoning. PCA is mainly applied to partitioning a universe repeatedly until all the partitioned sets are singular or indecomposable. Then we collect all the partitioned clusters as the granular knowledge and then use this knowledge to approximate the target set.
Optimized Artificial Neural Network Using Differential Evolution for Prediction of RF Power in
VHF/UHF TV and GSM 900 Bands for Cognitive Radio Networks
Sunday Iliya, Eric Goodyer, Mario Gongora and Jethro Shell
Centre for Computational Intelligence, School of Computer Science and Informatics, De
Montfort University, The Gateway, Leicester LE1 9BH, England, United Kingdom
John Gow
School of Engineering, Media and Sustainable Development, De Montfort University, The
Gateway, Leicester LE1 9BH, England, United Kingdom
Abstract —Cognitive radio (CR) technology has emerged as a promising solution to many wireless communication problems including spectrum scarcity and underutilization. The knowledge of Radio Frequency (RF) power (primary signals and/ or interfering signals plus noise) in the channels to be exploited by CR is of paramount importance, not just the existence or absence of primary users. If a channel is known to be noisy, even in the absence of primary users, using such channels will demand large quantities of radio resources (transmission power, bandwidth, etc) in order to deliver an acceptable quality of service to users. Computational Intelligence (CI) techniques can be applied to these scenarios to predict the required RF power in the available channels to achieve optimum
Quality of Service (QoS). While most of the prediction schemes are based on the determination of spectrum holes, those designed for power prediction use known radio parameters such as signal to noise ratio (SNR), bandwidth, and bit error rate. Some of these parameters may not be available or known to cognitive users. In this paper, we developed a time domain based optimized Artificial Neural Network (ANN) model for the prediction of real world RF power within the GSM 900, Very High Frequency (VHF) and
Ultra High Frequency (UHF) TV bands. The application of the models produced was found to increase the robustness of CR applications, specifically where the CR had no prior knowledge of the RF power related parameters. The models used implemented a novel and innovative initial weight optimi ation of the ANN’s through the use of differential evolutionary algorithms. This was found to enhance the accuracy and generalization of the approach.
Truss Topology Optimization with Species Conserving Genetic Algorithm
Jian-Ping Li and Felician Campean
School of Engineering and Informatics, University of Bradford, Bradford, UK.
Abstract — This paper is to apply the species conserving genetic algorithm (SCGA) to search multiple solutions of truss topology optimization problems in a single run. A realvector is used to represent the corresponding cross-sectional areas and a member is thought to be existent if its area is bigger than a critical area. A finite element analysis
29
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) model has been developed to deal with more practical considerations in modeling, such as existences of members, kinematic stability analysis and the computation of stresses and displacements. Cross-sectional areas and node connections are taken as decision variables and optimized simultaneously to minimize the total weight of trusses. Numerical results demonstrate that some truss topology optimization examples have many global and local solutions and different topologies can be found by using the proposed algorithm in a single run and some trusses have smaller weight than the solutions in the literature.
The Effect of Attribute Pairings in Intrusion Detection
Michael Milliken, Yaxin Bi and Leo Galway
School of Computing and Mathematics, University of Ulster, Jordanstown campus,
Newtownabbey, County Antrim BT37 0QB, United Kingdom
Abstract
—As Network Intrusions have become larger and more pervasive the methods of detection have changed, a number of systems use ensemble methods to improve upon results from single classifiers or algorithms. The solutions proposed in the literature achieve good results, which primarily focus on classification of Network Intrusions by tailoring classification algorithms and feature selection. However fewer studies focus on investigation of relation between pairs of attributes, such as IP address and Port, as a single attribute. This paper proposes an effect analysis of pairs of attributes in order to improve intrusion detection using an ensemble-based classification approach.
Towards Automated Monitoring of Adult Zebrafish
Qussay Al-Jubouri, Waleed Al-Nuaimy, Hamza S. AlZu'bi
Department of Electrical Engineering and Electronics, University of Liverpool, Liverpool
L69 3GJ, UK
Osama Zahran
Department of Electronic and Communication, Faculty of Electronic Engineering, Menofia
University, Egypt
Jonathan Buckley
Institute of Integrative Biology, University of Liverpool, Liverpool L69 3GJ, UK
Abstract
—Over the last two decades, zebrafish (Danio rerio) have emerged as an efficient model to aid in the research of a broad range of human diseases as well as such diverse applications as environmental modelling and drug discovery. Economically, the large number, low price and low maintenance requirements of this fish species encouraged its use for research. In addition to this, the study of zebrafish is being used to improve the understanding of fish physiology, with implications for fish welfare. In order to thoroughly model the behaviour, development and growth of these fish, it is important to be able to scrutinise the characteristics of individual fish as they respond to a range of stimuli, and to this end off-line fish recognition and on-line tracking using video data is employed.
Tracking and identifying such small and fast-moving objects is a challenge, and this paper seeks to address this using a behavioural analysis approach. Utilising single high resolution camera and two low-cost synchronised video cameras, the proposed systems captures front (face) and side (profile) pictures of each isolated fish as they swim past a given marker. The acquired images are then subject to three separate processing routes in order to satisfy three complementary but distinct objectives. Initially, fish face and profile features are extracted to aid the identification of individual fish. Then, for each fish identified, behavioural features such as the frequency and intensity of the operculum beat rate or breathing cycle are quantified in order to assess aspects of the fish welfare.
Additionally, the volume of each fish is estimated based on its profile dimensions, enabling the weight of the fish to be monitored throughout its lifetime. This paper presents
30
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) preliminary experimental considerations and findings of this on-going research project.
Results to date have been both encouraging and promising, validating the approach and the experimental configuration adopted.
Improving Motion State Change Object Detection by Using Block Background Context
Dazhen Lin, Donglin Cao and Hualin Zeng
Cognitive Science Department, ShenZhen Research Institute of Xiamen University, Fujian
Key Laboratory of the Brain-like Intelligent Systems, Xiamen University, Xiamen, China
Abstract
—Motion state change object detection, such as stopped objects detection, is one of important topics in Video Surveillance Systems. Generally, backgrounds in the most
Video Surveillance Systems have the property of pureness and selfsimilarity.
In this paper, we propose a block background context based background model to solve the motion state change problem. Unlike the classical background model, our approach first models blocks of background, and then determines the learning rate of each block background model by using the block background context information. There are two main advantages. First, the model adaptively selects the learning rate for each block of background model, and that is more flexible than the adaptive learning rate for the whole background. Second, context information helps the determination of true foreground and brings in more reliable information in foreground detection. Our experiments results show that our model outperforms the higher and lower learning rate Gaussian mixture background model in motion state change object detection.
Session 6: Intelligent Technologies for Healthcare and Behaviour Analysis 1
An Efficient System For Preprocessing Confocal Corneal Images For Subsequent Analysis
Mhd Saeed Sharif
Arun Brahma c a
, Rami Qahwaji a
, Sofyan Hayajneh b
, Stan Ipson a
, Rania Alzubaidi a
, a School of Electrical Engineering and Computer Science, University of Bradford,
Bradford, UK. b Isra University, Engineering Faculty, Amman, Jordan. c Manchester Royal Eye Hospital, Central Manchester University Hospitals NHS
Foundation Trust, Manchester Academic Health Science Centre, Manchester, M13 9WL,
UK.
Abstract — A confocal microscope provides a sequence of images of the various corneal layers and structures at different depths from which medical clinicians can extract clinical information on the state of health of the patient’s cornea. Preprocessing the confocal corneal images to make them suitable for analysis is very challenging due the nature of these images and the amount of the noise present in them. This paper presents an efficient preprocessing approach for confocal corneal images consisting of three main steps including enhancement, binarisation and refinement. Improved visualisation, cell counts and measurements of cell properties have been achieved through this system and an interactive graphical user interface has been developed.
Human Activity Classification Using A Single Accelerometer
Hamzah S. AlZubi
1
, Simon Gerrard-Longworth
2
, Waleed Al-Nuaimy
1
, Yannis Goulermas
1 and Stephen Preece
2
1 Department of Electrical Engineering and Electronics, University of Liverpool, Brownlow
Hill, Liverpool, L69 3GJ, UK
2 Centre for Rehabilitation and Human Performance Research, University of Salford
Fredrick Road Campus, Salford, M6 6PU, UK
31
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014)
Abstract
—Human activity recognition is an area of growing interest facilitated by the current revolution in body-worn sensors. Activity recognition allows applications to construct activity profiles for each subject which could be used effectively for healthcare and safety applications. Automated human activity recognition systems face several challenges such as number of sensors, sensor precision, gait style differences, and others.
This work proposes a machine learning system to automatically recognise human activities based on a single body-worn accelerometer. The in-house collected dataset contains 3D acceleration of 50 subjects performing 10 different activities. The dataset was produced to ensure robustness and prevent subject-biased results. The feature vector is derived from simple statistical features. The proposed method benefits from RGB-to-YIQ colour space transform as kernel to transform the feature vector into more discriminable features. The classification technique is based on an adaptive boosting ensemble classifier. The proposed system shows consistent classification performance up to 95% accuracy among the 50 subjects.
Multiple sensor outputs and computational intelligence towards estimating state and speed for control of lower limb prostheses
Pamela A. Hardaker, Benjamin N. Passow and David Elizondo
De Montfort University’s Interdisciplinary Group in Intelligent Transport Systems (DIGITS)
& Centre for Computational Intelligence (CCI), De Montfort University, Leicester, LE1 9BH,
United Kingdom
Abstract —For as long as people have been able to survive limb threatening injuries prostheses have been created. Modern lower limb prostheses are primarily controlled by adjusting the amount of damping in the knee to bend in a suitable manner for walking and running. Often the choice of walking state or running state has to be controlled manually by pressing a button. This paper examines how this control could be improved using sensors attached tofa the limbs of two volunteers. The signals from the sensors had features extracted which were passed through a computational intelligence system. The system was used to determine whether the volunteer was walking or running and their movement speed. Two new features are presented which identify the movement states of standing, walking and running and the movement speed of the volunteer. The results suggest that the control of the prosthetic limb could be improved.
Session 7: Intelligent Technologies for Healthcare and Behaviour Analysis 2
Blood-Glucose Pattern Mining Algorithm for Decision Support in Diabetes Management
Majid A. Al-Taee, Suhail N. Abood
School of Computing and Information Systems, Kingston University, UK
Waleed Al-Nuaimy
Department of Electrical Engineering and Electronics, University of Liverpool, UK
Ahmad M. Al-Taee
School of Medicine, The University of Jordan, Amman, Jordan
Abstract
—Pattern recognition has been an effective approach to identifying glycaemic patterns within self-monitored blood glucose (BG) data in diabetes mellitus patients. This paper presents a new BG pattern mining algorithm for more targeted therapeutic decision support in diabetes self-management. Based on patients’ BG readings which are collected via a handheld device and logged on a web-based health portal, the existing BG patterns are extracted in real-time and fed back to the patient along with appropriate therapeutic recommendations, educational modules and health care advice. The identified patterns help patients improve their blood glucose management and education about diabetes and
32
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) its complications. A functional prototype of the proposed system is developed and its endto-end functionality is successfully demonstrated. A pilot clinical study demonstrated positive user acceptability and interest in its decision support attributes for diabetes selfmanagement, making this a promising avenue for further research.
Integrated Microarray Analytics for the Discovery of Gene Signatures for Triple-Negative Breast
Cancer
Masood U.H. Zaka, Yonghong Peng
Department of Computing, School of Engineering and Informatics, University of Bradford,
Bradford, BD7 1DP, UK
Chris W. Sutton
Institute of Cancer Therapeutics, School of Life Sciences, University of Bradford, Bradford,
BD7 1DP, UK
Abstract
—Triple-negative breast cancers (TNBC) are clinically heterogeneous, an aggressive form of breast cancer with poor diagnosis and highly therapeutic resistant. It is urgently needed for identifying novel biomarkers with increased sensitivity and specificity for early detection and personalized therapeutic intervention. Microarray profiling offered significant advances in molecular classification but sample scarcity and cohort heterogeneity remains challenging areas. Here, we investigated diagnostics signatures derived from human triplenegative tissue. We applied REMARK criteria for searching for relevant studies and compared the eligible signatures by assessing their classification power for diagnosis using leave-one-out crossvalidation.
The cross-validation shows excellent classification accuracy ratios achieved for diagnosis using all the data sets. A subset signature (17-gene) extracted from the convergence of eligible signatures also achieved good classification accuracy of 89.37% across all data sets. We also applied gene ontology functional enrichment analysis to extract potentially common biological process, pathways and network involved in TNBC disease progression.
Through functional analysis, we recognized that these independent signatures have displayed commonalities in functional pathways of cell signaling, which play important role in the development and progression of TNBC. We have also identified five unique TNBC pathways genes (SYNCRIP, NFIB, RGS4, UGCG, LOX and NNMT), which could be important for therapeutic interventions as indicated by their close association with known drivers of TNBC and previously published experimental studies.
Prediction of Driver Fatigue: Approaches and Open Challenges
Hilal Abbood, Waleed Al-Nuaimy, Ali Al-Ataby, Sameh A. Salem and Hamzah S. AlZubi
Department of Electrical Engineering and Electronics, University of Liverpool, Brownlow
Hill, Liverpool, L69 3GJ, UK
Abstract
—Fatigue is a mental process that grows gradually and affects human reaction time and the consciousness. It is one of the causes of road fatal accidents around the globe. Although it is now generally accepted that fatigue plays an important role in road safety, it is still largely left to individual drivers to manage. The recent research in this area focuses on fatigue detection and the existing systems alert the drivers in severe fatigued stage. These systems use either physiological signs of the fatigue or the behavioural reaction to generate alerts. This research investigates the feasibility of using a group of fatigue symptoms (such as pupil response, gaze patterns, steering reaction and EEG) to build a robust fatigue detection algorithm that can be used in a real life system for the early prediction and avoidance of fatigue development. Intensive testing and validation stages are required to ensure the reliability and the suitability of the system that should be able to detect fatigue levels at different degrees of tiredness. Moreover, the proposed system
33
2014 14 th
UK Workshop on Computational Intelligence (UKCI2014) predicts subsequent stages of fatigue and generates an approximate behavioural model for each individual driver to enable more personalized and effective intervention.
Using Computational Methods for the Prediction of Drug Vehicles
Pritesh Mistry, Anna Palczewska, Daniel Neagu and Paul Trundle
School of Engineering and Computer Science, University of Bradford, UK, BD7 1DP
Abstract —Drug vehicles are chemical carriers that aid a drug’s passage through an organism. Whilst they possess no intrinsic efficacy they are designed to achieve desirable characteristics which can include improving a drug’s permea ility and or solu ility targeting a drug to a specific site or reducing a drug’s toxicity. All of which are ideally achieved without compromising the efficacy of the drug. Whilst the majority of drug vehicle research is focused on the solubility and permeability issues of a drug, significant progress has been made on using vehicles for toxicity reduction. Achieving this can enable safer and more effective use of a potent drug against diseases such as cancer. From a molecular perspective, drugs activate or deactivate biochemical pathways through interactions with cellular macromolecules resulting in toxicity. For newly developed drugs such pathways are not always clearly understood but toxicity endpoints are still required as part of a drug’s registration. An understanding of which vehicles could be used to ameliorate the unwanted toxicities of newly developed drugs would be highly desirable to the pharmaceutical industry. In this paper we demonstrate the use of different classifiers as a means to select vehicles best suited to avert a drug’s toxic effects when no other information a out a drug’s characteristics is known. Through analysis of data acquired from the Developmental Therapeutics Program (DTP) we are able to establish a link
etween a drug’s toxicity and vehicle used. We demonstrate that classification and selection of the appropriate vehicle can be made based on the similarity of drug choice.
34
